spark-llm-finetune-tj

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>

<details><summary>See axolotl config</summary>

axolotl version: 0.9.2

base_model: pretrained_models/Spark-TTS-0.5B/LLM
# Automatically upload checkpoint and final model to HF
hub_model_id: muhtasham/spark-llm-finetune-tj

trust_remote_code: true

strict: false

datasets:
  - path: data/output_prompt.jsonl
    type: completion

dataset_prepared_path:
val_set_size: 0.05
output_dir: ./outputs/out


sequence_len: 4098
sample_packing: true
eval_sample_packing: true
pad_to_sequence_len: true

wandb_project: spark-tts
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 4
num_epochs: 50
optimizer: adamw_torch_fused
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: true

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 50
xformers_attention:
flash_attention: true

warmup_steps: 10
evals_per_epoch: 1
save_steps: 5000
debug:
deepspeed:
weight_decay: 0.0

</details><br>

spark-llm-finetune-tj

This model was trained from scratch on the data/output_prompt.jsonl dataset. It achieves the following results on the evaluation set:

• Loss: 5.2546

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 50.0

Training results

Framework versions

• Transformers 4.51.3
• Pytorch 2.7.1+cu126
• Datasets 3.5.1
• Tokenizers 0.21.1